Unicast IPv4 Routing Tools and Settings

Applies To: Windows Server 2003, Windows Server 2003 R2, Windows Server 2003 with SP1, Windows Server 2003 with SP2

In this section

  • Unicast IPv4 Routing Tools

  • Unicast IPv4 Routing Registry Entries

  • Related Information

You can use the unicast IPv4 routing tools and registry settings described here to enable, configure, and manage the unicast IPv4 routing features provided by the Microsoft® Windows® Server™ 2003 Routing and Remote Access service.

Unicast IPv4 Routing Tools

The sets of tools listed in the following table are associated with unicast IPv4 routing as supported by the Windows Server 2003 Routing and Remote Access service.

Available Tools for Unicast IPv4 Routing

Category Tools in This Set

Command-line tools used for unicast IPv4 routing

  • Netsh.exe, including:

    • Netsh ras commands for file tracing

    • Netsh routing IP commands for IP packet filtering and for configuring interfaces, persistent and non-persistent static routes, and preference levels for routing protocols (route sources)

    • Netsh routing IP OSPF commands

    • Netsh routing IP relay commands (for the DHCP relay agent)

    • Netsh IP RIP commands

    • Netsh routing IP routerdiscovery commands (for ICMP router discovery)

  • Pathping.exe

  • Ping.exe

  • Route.exe

  • Tracert.exe

Graphical user interface tools used for unicast IPv4 routing

  • Network Connections (default gateway options only)

  • Rrasmgt.msc — Routing and Remote Access snap-in, including:

    • Options for enabling, disabling, or refreshing the Routing and Remote Access service

    • General options (event logging and route source preference options)

    • RIP for IP options

    • OSPF options

    • DHCP Relay Agent options

    • IP packet filtering options

    • ICMP router discovery options

Each of these tools is described in the following sections.

You can specify commands for the command-line tools for one or more remote servers by establishing a remote access connection to each server from a workstation running Microsoft Windows XP or from another server, and then opening a command prompt window in the remote access window for each server. You can also specify Network Connections and Routing and Remote Access snap-in options by establishing a remote access connection. In addition, you can use the Remote Access snap-in to manage multiple servers running the Routing and Remote Access service by accessing the node for each server.

Netsh.exe: Netsh Commands for Managing IPv4 Routing

Category The Netsh commands for IPv4 routing are a subset of the Netsh command-line tools that are included with the Windows Server 2003 operating system. In many cases, a Netsh command can be used to manage the Routing and Remote Access service as an alternative to using an option available in the Routing and Remote Access snap-in. Netsh also includes commands for IPv6.

Version compatibility The Netsh commands for IP routing are compatible with the Windows Server 2003 operating system. Netsh commands were first introduced in Microsoft Windows 2000 Server and were expanded to include additional commands in Windows Server 2003.

The Netsh commands help network administrators manage a TCP/IP network. You can use Netsh locally or remotely to display or modify the configuration of services or protocols on Windows-based computers. The Netsh command-line interface is scriptable, which lets you perform batch configurations or remote network administration from a single location.

Netsh provides sets of commands (also known as contexts) for performing a wide range of network configuration tasks. In addition to a few commands from the Netsh ras and Netsh interface contexts, this document describes several Netsh Routing IP contexts used in conjunction with the unicast IPv4 routing features of the Routing and Remote Access service.

For more information about Netsh, see Command-Line Reference for Windows Server 2003, Standard Edition. In particular, for information about IP routing-related commands not included in this technical reference, see the following Netsh contexts:

  • Netsh ras context, for remote access configuration.

  • Netsh interface context, for configuration of demand-dial interfaces.

  • Netsh routing IP NAT context, for network address translation (NAT).

  • Netsh interface IPv6 context, for manually configuring IPv6 settings.

  • Netsh routing IP IGMP context, for multicast routing.

Each of the listed Netsh contexts contains unicast routing–related commands not included in this technical reference, except for the IGMP context, which contains multicast routing–related commands.

How to access Netsh commands

You can access a particular command in any Netsh context in one of the following two ways:

Type the entire path for the command at the command prompt. For example, to view IPv4 router discovery configuration data, type the following command at a command prompt:

C:\>netsh routing ip routerdiscovery show interface name="Local Area Connection"

The preceding command produces the following output:

Router Discovery Information for Local Area Connection 3

----------------------------------------------------------------------

Router Discovery is: Disabled

Minimum Advertisement Interval: 7 minutes

Maximum Advertisement Interval: 10 minutes

Advertisement Lifetime: 30 minutes

Preference Level: 0

Access the context containing the command that you want to use, and then type only the specific command. The following series of commands displays the same output as the preceding example:

C:\>netsh

netsh>routing

netsh routing>ip

netsh routing ip>routerdiscovery

netsh routing ip routerdiscovery>show interface name="Local Area Connection"

The preceding series of commands produce the following output:

Router Discovery Information for Local Area Connection 3

----------------------------------------------------------------------

Router Discovery is: Disabled

Minimum Advertisement Interval: 7 minutes

Maximum Advertisement Interval: 10 minutes

Advertisement Lifetime: 30 minutes

Preference Level: 0

The following sections contain a brief description of unicast IPv4 routing commands that are related to the Routing and Remote Access service and that are available for the following Netsh contexts:

  • Netsh ras context (includes only commands for configuring file tracing)

  • Netsh routing IP context (includes only commands for configuring IP packet filtering, interfaces, persistent and non-persistent static routes, and preference levels for route sources)

  • Netsh routing IP OSPF context

  • Netsh routing IP relay context

  • Netsh routing IP RIP context

  • Netsh routing IP routerdiscovery context

Netsh ras context

The Routing and Remote Access service provides extensive tracing capability to help troubleshoot complex network problems. Separate routing components can be independently enabled so that tracing information can be logged to files (known as file tracing). Logged data includes information about internal component variables, function calls, and interactions. You can enable and disable tracing for routing protocols while the router is running.

Note

You can also enable file tracing for Routing and Remote Access components, such as the RIP and OSPF routing protocols, by editing the registry (see Unicast IPv4 Routing Registry Entries later in this document). However, using Netsh commands is the recommended method.

Tracing consumes system resources and must be used sparingly. After you obtain the trace or identify the problem, immediately disable tracing. In particular, do not leave tracing enabled on multiprocessor computers.

The Netsh ras context includes primarily remote access–related commands, which are not included here. The following table lists only the file tracing commands for the RIP v2 and OSPF routing protocols that are available in the Netsh ras context.

Unicast IPv4 Routing Commands in the Netsh Ras Context

Command Task

Set

You can use the following commands:

  • Set tracing component enable. Enables file tracing for the specified component. For example, set tracing iprip2 enable or set tracing ospf enable.

  • Set tracing component disable. Disables file tracing for the specified component. For example, set tracing iprip2 disable or set tracing ospf disable.

  • Set tracing * enable. Enables file tracing on all components.

Show

You can use the following commands:

  • Show tracing. Lists components for which it is possible to enable file tracing and shows whether file tracing for each one is currently enabled or disabled.

  • Show tracing component. Shows whether file tracing for the specific component is enabled. For example, show tracing iprip2 or show tracing ospf.

Most of the commands available in the Netsh ras context are used for remote access networking, such as for demand-dial or virtual private network (VPN) connections. These commands are also unicast IPv4 commands, but they are not covered in this technical reference. For more information about tools for remote access networking, see Demand Dial Tools and Settings and VPN Tools and Settings.

Netsh routing IP context

The partial set of commands included in this document for this context are the Netsh commands used to configure IP packet filtering, which can help increase security on a private IP internetwork connected to the Internet by specifying which IP packet traffic is accepted or rejected; the commands used to configure interfaces, persistent and non-persistent static routes; and preference levels for route sources. Also included, in separate sections later in this document, are several relevant contexts accessible under the Netsh routing IP context: the OSPF, RIP, relay (for the DHCP relay agent), and IP router discovery contexts.

Excluded from this technical reference are Netsh commands used for IPv4 multicast routing, network address translation (NAT), DHCP, DNS, and remote access.

The following table describes commands available in the Netsh routing IP context that are relevant to the unicast IPv4 routing services provided by the Routing and Remote Access service.

Unicast IPv4 Routing Commands in the Netsh Routing IP Context

Command Task

Add

  • Add filter. Adds an IP packet filter to the specified interface and configures one or more of the following parameters:

    • The name of the interface on which to add the filter.

    • The type of filter (input filter, output filter, or a filter for a dial-up interface).

    • The source IP address and subnet mask of the packet to be filtered.

    • The destination IP address and subnet mask of the packet to be filtered.

    • The protocol type (TCP, TCP-EST, or UDP) and the source and destination ports for the packet to be filtered.

    • TCP-EST refers to TCP [established] and is used, for example, in a site-to-site VPN or dial-up connection when a calling router connects to an answering router. If you specify TCP [established], traffic is accepted only if the calling router has already initiated the TCP connection.

    • The ICMP type and ICMP code of the packet to be filtered.

  • Add interface. Enables IP forwarding on an interface as it adds the interface (add interface name=“Interface Name” state=enable) or disables IP forwarding on an interface as it adds the interface (add interface name= “Interface Name” state=disable).

  • Add persistentroute. Adds a persistent static route to the specified interface, including:

    • The destination IP address and subnet mask for the route.

    • The name of the interface on which the route is to be added.

    • The next hop (gateway) IP address for the route (omit for point-to-point interfaces).

    • The connection type for this route, that is, static (which triggers a demand-dial connection) or nondod (which does not trigger a demand-dial connection).

    • The preference for this route, and its metric. The metric is used to break ties among routes with equal preference.

    • The view (unicast, multicast, or both) for this route.

  • Add preferenceforprotocol. Sets the preference level for any of the following route sources:

    • Autostatic

    • Local

    • NetMgmt (network management, such as SNMP)

    • Nondod (a static route assigned to a LAN interface)

    • OSPF

    • RIP

    • Static (a static route assigned to a demand-dial interface)

    noteNote
    If two routes exist to the same destination, the route learned by the route source with a higher preference is preferred over the route learned by the route source with a lower preference regardless of the route metric. For example, if the preference level for OSPF is higher than the preference level for RIP v2, an OSPF route with a metric of 5 is entered into the routing table rather than the same route discovered by RIP with a metric of 3.
  • Add rtmroute. Adds a non-persistent route used for network management to the specified interface name, including:

    • The destination IPaddress and subnet mask for the route.

    • The name of the interface on which the route is to be added.

    • The next hop (gateway) IP address for the route (omit for point-to-point interfaces).

    • The preference for this route, and its metric. The metric is used to break ties among routes with equal preference).

    • The view (unicast, multicast, or both) for this route.

Delete

  • Delete filter. Deletes an IP packet filter from the specified interface.

  • Delete interface. Deletes the specified interface, which also removes IP forwarding on that interface.

  • Delete persistentroute. Deletes a persistent static route from the specified interface.

  • Delete preferenceforprotocol. Deletes the preference level for a specified routing protocol type (also known as route source).

  • Delete rtmroute. Deletes a non-persistent route used for network management when either the interface name or the next hop count is specified.

Dump

Creates a script that contains the current configuration for this server. If you save the script to a file, you can use it to restore altered configuration settings or to configure a new server.

“Current configuration” refers to all configuration settings in and below the specific node in the Routing and Remote Access snap-in at which the command “dump” is given. Because this table refers to the Netsh routing IP context, in this case, the dump command includes all configuration settings in the IP Routing node and any configured child nodes, including the General, Static Routes, DHCP Relay Agent, OSPF, NAT/Basic Firewall, and IGMP nodes.

If you plan to make extensive changes to the configuration of this server, a recommended practice is to begin the configuration session by using the dump command before you start, in case you need to restore the configuration later.

Set

  • Set filter. Changes the default action (drop or forward) for a filter type (input, output, or dial-up) or changes the fragmentation check setting (enable or disable) on the specified interface.

  • Set interface. Enables or disables the specified interface.

  • Set loglevel. Sets the global logging level to one of the following: none; errors only; errors and warnings; or errors, warnings, and informational messages.

  • Set persistentroute. Modifies a persistent static route on the specified interface by changing one or more of the following:

    • The destination IP address and subnet mask for the specified route.

    • The name of the interface on which the route is to be modified.

    • The next hop for the route (omit for point-to-point interfaces).

    • The connection type for this route that is static (which triggers a demand-dial connection) or nondod (which does not trigger a demand-dial connection).

    • The preference for the route.

    • The metric for the route. The metric is used to break ties among routes with equal preference.

    • Whether the route is valid only for unicast traffic, only for multicast traffic, or for both types of traffic.

  • Set preferenceforprotocol. Sets the preference level for any of the following protocol types (also known as route sources):

    • Autostatic

    • Local

    • NetMgmt (network management, such as SNMP)

    • Nondod (a static route assigned to a LAN interface)

    • OSPF

    • RIP

    • Static (a static route assigned to a demand-dial interface)

  • Set rtmroute. Modifies a non-persistent route used for network management on the specified interface:

    • The destination IP address and subnet mask for the route.

    • The interface on which the route is to be added.

    • The next hop (gateway) IP address for the route (omit for point-to-point interfaces).

    • The preference for this route.

    • The metric for this route. The metric is used to break ties among routes with equal preference.

    • The view (unicast, multicast, or both) for this route.

Show

  • Show filter. For all interfaces or for the specified interface, indicates whether input, output, or demand-dial IP packet filters are configured and shows whether fragmentation checking is enabled. If a filter exists for a given interface, displays filter information for that interface.

  • Show interface. For all interfaces or for the specified interface, displays information including interface type, vendor, routing protocols running on a given interface, and whether or not an interface is enabled.

  • Show loglevel. Displays the global logging level, which is one of the following: none; errors only; errors and warnings; or errors, warnings, and informational messages.

  • Show persistentroutes. For all interfaces or for the specified interface, displays persistent static routes, including:

    • Each route and its address prefix . The address prefix is shown in classless interdomain routing (CIDR) notation, for example, 10.0.0.1/8.

    • The protocol (route source) that added this route.

    • The preference for this route, and its metric.

    • The next hop (gateway) IP address for the route, and the interface that this route uses to forward packets.

    • The view (unicast, multicast, or both) for this route.

  • Show preferenceforprotocol. Lists the current preference level (relative priority) for each IP routing protocol type (each type of route source).

  • Show protocol. Displays all configured IP protocols, including unicast IP protocols.

  • Show rtmdestinations1. Displays the IP forwarding table. IP uses the forwarding table to forward packets.

  • Show rtmroutes1. Displays the IP routing table. IP uses the routing table to store all routes from all possible sources.

1. The Routing and Remote Access service on a router running Windows Server 2003 maintains the IP routing table by using a component called the Route Table Manager, abbreviated as “rtm” in the commands rtmroutes and rtmdestinations. The Route Table Manager updates the IP forwarding table based on incoming route information from multiple sources.

Netsh routing IP OSPF context

Open Shortest Path First (OSPF), the most common routing protocol used on large and very large IPv4 internetworks, enables the dynamic exchange of IPv4 routing information between OSPF routers across complex internetworks. Typically, OSPF is deployed on private IP internetworks connected to the public Internet, including internetworks that span multiple geographical regions, nations, or continents.

OSPF is structured as an autonomous system made up of one or more OSPF areas. Each area is made up of a logical collection of networks, routers, and links, and each area includes special OSPF routers that enable communication among the constituent areas. OSPF routers in each area dynamically build and synchronize a database of the OSPF network topology, called a link state database (LSDB). The LSDB is a topological map of all routers in the same OSPF area, and OSPF routers can update the LSDB in response to topology changes.

The following table describes the commands available in the Netsh routing IP OSPF context. The OSPF routing protocol is provided by the Windows Server 2003 Routing and Remote Access service.

Commands in the Netsh Routing IP OSPF Context

Command Task

Add

  • Add area. Adds an OSPF area. Specify the area ID, which is a 32-bit number expressed in dotted decimal notation. The only required OSPF area is the OSPF backbone area, which requires the reserved area ID of 0.0.0.0. An OSPF backbone area is either the sole area in the OSPF autonomous system, or it is the central area to which all other OSPF areas in this OSPF autonomous system connect and through which packets are routed from one non-backbone area to another non-backbone area.

    Although an OSPF area ID is neither an IP address nor an IP network ID, if this is not the backbone area and if the entire area can be summarized with a single network ID, you can set the area ID to match the single network ID. For example, if an area contains all of the subnets for the IP network 10.1.0.0/16, you can set the area ID to 10.1.0.0.

  • Add interface. Enables OSPF on the specified interface.

  • Add neighbor. Adds an OSPF neighbor router on a non-broadcast multiple access (NBMA) network. Specify the IP address of the OSPF router at the opposite end of the NBMA link from this router, and then specify the router priority of the neighbor router.

  • Add protofilter. Adds an OSPF protocol filter (autostatic, local, RIP, SNMP, non-demand-dial, or static).

  • Add range. Adds the network ID and subnet mask pair that define a range of addresses for the specified area ID. This {network ID, subnet mask} pair is known as a summary route and is used to summarize routes within this OSPF area.

  • Add routefilter. Adds an OSPF route filter. Specify the IP address and subnet mask pair of the OSPF route to be filtered.

  • Add virtif. Adds a virtual interface (also called a virtual link). Specify the OSPF transit area ID and the OSPF router ID of the virtual OSPF neighbor router. A virtual link is a logical connection that enables the exchange of routing information between a backbone area border router (ABR) and an ABR that is not physically connected to the backbone area across a transit area. You cannot configure a virtual link across the backbone area.

Delete

  • Delete area. Deletes the specified OSPF area by deleting its area ID.

  • Delete interface. Disables OSPF on the specified interface.

  • Delete neighbor. Deletes the specified OSPF neighbor router on an NBMA network.

  • Delete protofilter. Deletes the specified OSPF protocol filter (autostatic, local, RIP, SNMP, non-demand-dial, or static).

  • Delete range. Deletes the IP address and subnet mask pair that define a range of addresses (a summary route) for the specified OSPF area ID.

  • Delete routefilter. Deletes the specified OSPF route filter.

  • Delete virtif. Deletes the specified virtual interface on this OSPF router.

Dump

Creates a script that contains the current configuration for OSPF. If you save the script to a file, you can use it to restore altered configuration settings or to configure a new OSPF router.

If you plan to make extensive changes to the configuration of an OSPF router, a recommended practice is to begin the configuration session by using the dump command before you start, in case you need to restore the configuration later.

Install

Installs the OSPF routing protocol under the IP routable protocol.

Set

  • Set area. For the specified OSPF area, sets the following:

    • Whether plaintext passwords are required. Plaintext passwords are used by routers to identify each other, not for security. Passwords are enabled by default, and all interfaces in the same area must use the same password.

    • Whether this area is a stub area, and, if so, the cost of the summary default route that an OSPF router in this area uses to reach all destinations external to its OSPF autonomous system.

    • Whether the OSPF area imports summary advertisements.

  • Set global. For the specified OSPF router, sets the following:

    • Whether the router acts as an autonomous system boundary router (ASBR).

    • Whether the amount of information logged is one of the following: none; errors only; errors and warnings; or errors, warnings, and informational messages.

  • Set interface. For the specified OSPF interface, sets the following:

    • The name, IP address, and subnet mask of the interface; whether the interface is enabled or disabled; and the OSPF area ID.

    • Whether the interface is on an NBMA, point-to-point, or broadcast network, and the interface priority.

    • The estimated number of seconds to transmit a link state update packet over this interface; the interval in seconds between link state advertisement retransmissions for adjacencies; the interval in seconds between transmissions of hello packets; the number of seconds before a neighboring router considers this router to be down; and (for an NBMA interface only) the interval in seconds between OSPF network polls.

    • The cost of sending a packet from this interface.

    • The password (if simple password authentication is enabled for this OSPF area). Passwords are enabled by default, and all interfaces in the same area must use the same password.

    • The maximum size, in bytes, of IP packets carrying OSPF information that can be sent without fragmentation.

  • Set protofilter. For the OSPF protocol filter (autostatic, local, RIP, SNMP, non-demand-dial, or static), specifies whether the action for IP packets is drop or accept.

  • Set routefilter. For the OSPF route filter, specifies whether IP packets are accepted or dropped.

  • Set virtif. For the specified OSPF virtual interface (virtual link), sets the following:

    • The IP address of the transit area, and the OSPF router ID of the virtual neighbor router.

    • The estimated number of seconds to transmit a link state update packet over this interface; the interval in seconds between link state advertisement retransmissions for adjacencies; the interval in seconds between transmissions of hello packets; and the number of seconds before a neighboring router considers this router to be down.

    • The password (if simple password authentication is enabled).

Show

  • Show area. Shows configuration information about this OSPF area, including:

    • Whether the area is created and enabled as an OSPF area, or created but not enabled.

    • Authentication type (such as simple password authentication), if any. Plaintext passwords for this type of authentication are used by routers to identify each other, not for security.

    • Whether the area is a stub area, and if so, its metric.

    • Whether the area imports summary advertisements.

  • Show areastats. For each OSPF area, or for the specified area, shows the state (Up or Down) and the shortest path first (SPF) count. The SPF count is the number of least-cost paths that this router has calculated from itself to each other router and subnet in this OSPF autonomous system (if only one OSPF area exists) or to each other router in the local area (if multiple OSPF areas exist).

  • Show global. Shows global configuration information for this OSPF router:

    • Whether the router is created and enabled as an OSPF router, or created but not enabled.

    • The OSPF router ID.

    • Whether the router is an area border router (ABR).

    • The specified level of logging for this router.

  • Show interface. Shows OSPF configuration information for the specified OSPF interface, including:

    • Whether the interface is created and enabled as an OSPF interface, or created but not enabled.

    • The IP address, subnet mask, area ID, interface type, and router priority of the OSPF interface.

    • The current values for transit delay, hello interval, dead interval, poll interval, metric, and maximum transmission unit (MTU) size.

  • Show lsdb. Displays information about the OSPF link state database (LSDB):

    • Area. The identification number of the OSPF area for each link state advertisement (LSA) in the table.

    • Type. The type of advertisement, including Router, Net-Summary, Stub, Network, ASBR-Summary, or AS External.

    • Link State ID. The identification number of each advertisement shown in the output. The advertisements vary by advertisement type.

    • Adv Router. The OSPF router ID of the advertising router.

    • Age. The age (in seconds) since the advertisement originated.

    • Sequence. A sequence number that identifies old or duplicate advertisements.

  • Show neighbor. Lists OSPF neighbor routers and information about each neighbor:

    • Neighbor. The IP address of each neighbor router.

    • Neighbor ID. Each neighbor router’s OSPF router ID.

    • State. Displays one of the following: Down; Attempt; Init; 2-way; ExStart; Exchange; Loading; or Full. For a definition of each of these states, see “How OSPF Works” in How Unicast IPv4 Routing Protocols and Services Work.

    • Priority. The router priority of each neighboring router.

    • State-changes and errors. The total number of changes of state and errors.

    • Queue length. The length of the retransmit queue.

  • Show protofilter. For the OSPF protocol filter (autostatic, local, RIP, SNMP, non-demand-dial, or static), shows whether the action for IP packets is drop or accept.

  • Show routefilter. Displays OSPF route filter information.

  • Show virtif. Displays information about the specified virtual link.

  • Show virtifstats. Displays information about all OSPF virtual links:

    • Area ID. The area ID of the transit area for the virtual link.

    • Neighbor ID. The neighboring router’s OSPF router ID.

    • State. Displays one of the following: Down; Loopback; Waiting; Point-to-point; Designated-router; Backup designated-router; Other; or Unknown.

    • Sent packets. The number of packets sent over this virtual link.

    • Received packets. The number of packets received over this virtual link.

    • Dropped packets. The number of packets that failed to be transmitted over this virtual link.

Uninstall

Removes the OSPF routing protocol.

Netsh routing IP relay context

Any IPv4 subnet that contains DHCP clients requires either a DHCP server or a properly configured DHCP relay agent to provide address leases to those DHCP clients. A computer acting as a DHCP relay agent forwards DHCP messages between DHCP clients on an IP subnet that has no DHCP server and a DHCP server located on a different subnet. Using DHCP relay agents make it unnecessary to install a separate DHCP server on each subnet.

The following table describes the commands related to DHCP and DHCP relay agent in the Netsh routing IP relay context. The capability to enable a router to act as a DHCP relay agent is provided by the Windows Server 2003 Routing and Remote Access service.

Commands in the Netsh Routing IP Relay Context

Command Task

Add

  • Add dhcpserver. Adds a DHCP server to the global list of DHCP servers.

  • Add interface. Enables the DHCP Relay Agent service on the specified interface.

Delete

  • Delete dhcpserver. Deletes a DHCP server from the global list of DHCP servers.

  • Delete interface. Disables the DHCP Relay Agent service on the specified interface.

Dump

Creates a script that contains the current configuration for the DHCP relay agent. If you save the script to a file, you can use it to restore altered configuration information.

If you plan to make extensive changes to the configuration of a DHCP relay agent, a recommended practice is to begin the configuration session by using the dump command before you start, in case you need to restore the configuration later.

Install

Installs the DHCP Relay Agent service.

Set

  • Set global. Sets the logging level for a DHCP relay agent to one of the following: none; errors only; errors and warnings; or errors, warnings, and informational messages.

  • Set interface. On the specified interface, updates the following DHCP relay agent information:

    • Whether the DHCP Relay Agent service is enabled or disabled.

    • The maximum hop count before a DHCP packet is dropped.

    • The minimum number of seconds that the DHCP relay agent waits before forwarding a DHCP packet (from a DHCP client computer initiating an IP address renewal request) to a listed DHCP server.

    • The delay allows a DHCP server on the local subnet (if one is available) time in which to respond to the DHCP client’s request. If no DHCP server responds after the configured interval, the DHCP relay agent then processes the request.

Show

  • Show global. Displays the following information about the DHCP relay agent configured on this router:

    • The logging level: none; errors only; errors and warnings; or errors, warnings, and informational messages.

    • The maximum number of DHCP packets in the router’s queue (for all interfaces).

    • The number of DHCP servers in the global list.

  • Show ifbinding. For the specified interface, displays IP address bindings. If no interface is specified, displays IP address bindings for each interface on which the DHCP relay agent is enabled.

  • Show ifconfig. For the specified interface, displays:

    • Whether the DHCP relay agent is bound and enabled.

    • The maximum hop count before a DHCP packet is dropped.

    • The minimum number of seconds from startup before a DHCP packet is sent to a listed DHCP server.

    If no interface is specified, displays this information for each interface on which the DHCP relay agent is enabled.

  • Show ifstats. For the specified interface, displays:

    • The number of send and receive failures.

    • Address Resolution Protocol (ARP) update failures.

    • Requests and replies received or discarded.

    If no interface is specified, displays statistics for each interface on which the DHCP relay agent is enabled.

  • Show interface. For the specified interface, displays:

    • The state (enabled or disabled, and bound or unbound).

    • The relay mode (enabled or disabled).

    • The maximum hop count.

    • The minimum number of seconds that a DHCP relay agent waits after a DHCP client initiates an IP address renewal request before the DHCP relay agent forwards the DHCP client request to a remote DHCP server.

Uninstall

Removes the DHCP Relay Agent service.

Netsh routing IP RIP context

Routing Information Protocol (RIP), the primary dynamic routing protocol used on small or medium-sized IPv4 internetworks (10 to 50 networks), enables the dynamic exchange of IPv4 routing information between RIP routers. RIP routers can update IP internetwork topology changes as networks are added or removed and as links fail and are restored.

The following table describes the commands available in the Netsh routing IP RIP context. The RIP routing protocol is provided by the Windows Server 2003 Routing and Remote Access service, which supports both the updated version of RIP, RIP version 2 (RIP v2) and the earlier version of RIP, RIP version 1 (RIP v1).

Commands Available in the Netsh Routing IP RIP Context

Command Task

Add

  • Add acceptfilter. Adds an acceptance filter for routes received on an interface by specifying the IP address range.

  • Add announcefilter. Adds a filter for routes announced on an interface by specifying the IP address range.

  • Add interface. Enables RIP on the specified interface, and configures the following:

    • A metric value for routes based on the interface.

    • Whether to perform on-demand updates or periodic updates.

    • Whether to announce no routes, RIP v1 routes only, RIP v1 and v2 routes, or RIP v2 routes only.

    • Whether to accept no routes, RIP v1 routes only, RIP v1 and v2 routes, or RIP v2 routes only.

    • An interval for route expiration, route removal, and full update.

    • Whether simple password authorization is required. This type of authorization uses plaintext passwords, which are used by routers to identify each other, not for security.

    • The route tag.

    • Whether to use unicast mode in addition to the default mode, use only unicast mode, or to disable unicast mode. The default mode is to use broadcast or multicast only (but not unicast), based on the RIP version used (that is, RIP v1 broadcast, RIP v 2 broadcast, or RIP v 2 multicast).

    • Whether to include, exclude, or disable acceptance filters.

    • Whether to include, exclude, or disable announcement filters.

    • A plaintext password (1–16 characters).

  • Add neighbor. Adds a RIP neighbor router on an interface. Specify the IP address for the RIP neighbor router.

  • Add peerfilter. Adds a filter for another RIP router from which this router will accept announcements. Specify the IP address of the peer server. By default, RIP announcements from all sources are accepted. By configuring filters accepting announcements from one or more RIP peers, RIP announcements from unauthorized RIP routers are discarded.

Delete

  • Delete acceptfilter. Deletes an acceptance filter for routes received on an interface.

  • Delete announcefilter. Deletes an announcement filter set for an interface.

  • Delete interface. Disables RIP from the specified interface.

  • Delete neighbor. Deletes a RIP neighbor from an interface.

  • Delete peerfilter. Deletes a filter that specifies another RIP router from which this router will accept announcements.

Dump

Creates a script that contains the current configuration for RIP. If you save the script to a file, you can use it to restore altered configuration information or to configure a new RIP router.

If you plan to make extensive changes to the configuration of a RIP router, a recommended practice is to begin the configuration session by using the dump command before you start, in case you need to restore the configuration later.

Install

Installs the RIP routing protocol under the IP routable protocol.

Set

  • Set flags. Sets RIP-related flags for the specified interface by specifying one or more of the following values:

    • Clear. Clear all currently set RIP-related flags.

    • Splithorizon. Enable RIP for split horizon.

    • Poisonreverse. Enable RIP for poison reverse.

    • Triggeredupdates. Enable RIP triggered updates.

    • Cleanupupdates. Enable cleanup updates for RIP.

    • Accepthostroutes. Enable acceptance of host routes.

    • Acceptdefaultroutes. Enable acceptance of default routes.

    • Senddefaultroutes. Enable sending of default routes.

    • Nosubnetsummary. Disable subnet summarization.

  • Set global. Sets the following global RIP parameters:

    • Whether to set the level of logging for RIP events to one of the following: none; errors only; errors and warnings; or errors, warnings, and informational messages.

    • Minimum interval in minutes between triggered updates.

    • Whether to include peers, exclude peers, or disable peer mode.

  • Set interface. Modifies one or more of the following on the specified interface:

    • A metric value for routes based on this interface.

    • Whether to perform on-demand updates or periodic updates.

    • Whether to announce no routes, RIP v1 routes only, RIP v1 and v2 routes, or RIP v2 routes only.

    • Whether to accept no routes, RIP v1 routes only, RIP v1 and v2 routes, or RIP v2 routes only.

    • An interval for route expiration, route removal, and full update.

    • Whether simple password authorization is required. This type of authorization uses plaintext passwords, which are used by routers to identify each other, not for security.

    • The route tag.

    • Whether to also use unicast mode in addition to default mode, only use unicast mode, or to disable unicast mode. The default mode is to use broadcast or multicast only (but not unicast), based on the RIP version used (that is, RIP v1 broadcast, RIP v 2 broadcast, or RIP v 2 multicast).

    • Whether to include, exclude, or disable acceptance filters.

    • Whether to include, exclude, or disable announcement filters.

    • A plaintext password (16 characters maximum).

Show

  • Show flags. Displays RIP flags set on the specified interface.

  • Show global. Displays global parameters for this RIP router:

    • The specified level of logging: none; errors only; errors and warnings; or errors, warnings, and informational messages.

    • The maximum size for the receive queue and for the send queue.

    • The minimum interval in minutes between triggered updates.

    • The status (include, exclude, or disable) for peer filter mode.

    • The number of peer filters.

  • Show globalstats rr=n. Displays the number of route changes and the number of responses sent every n seconds. In this command, rr stands for refresh rate, that is, the interval, in seconds, that you specify after which the statistics for route changes and responses sent are refreshed.

  • Show ifbinding. Displays IP address bindings for the specified RIP interface.

  • Show ifstats. Displays RIP statistics for the specified RIP interface, including:

    • Send and receive failures.

    • Requests sent and received.

    • Responses sent and received.

    • Bad response packets and bad response entries received.

    • Triggered updates sent.

  • Show interface. Displays RIP configuration for the specified interface, including:

    • Metric.

    • Update, accept, and announce modes.

    • Interval for route expiration, route removal, and full update.

    • Authentication type and password.

    • Route tag.

    • Unicast peer mode, accept filter mode, and announce filter mode.

    • Unicast peer count, accept filter count, and announce filter count.

  • Show neighbor. Shows RIP neighbor statistics.

Uninstall

Removes the RIP routing protocol

Netsh routing IP routerdiscovery context

In unicast IPv4 routing, ICMP router discovery is an alternative to using manually configured or DHCP-configured default gateways.

Note

In IPv6 unicast routing, IPv6 Neighbor Discovery (ND) replaces and improves the functionality of Address Resolution Protocol (ARP), ICMP Router Discovery, and ICMP Redirect, which together are the IPv4 counterparts for IPv6 ND. ND, which is a set of messages and processes that determine relationships between neighboring nodes, is described in RFC 2461, "Neighbor Discovery for IP Version 6 (IPv6)," in the IETF RFC Database.

ICMP router discovery automates the discovery and configuration of the default gateway so that IP hosts can dynamically discover the best default gateway to use on a subnet and can automatically switch to another default gateway if the first default gateway fails or if the network administrator changes router preferences.

The following table describes the commands for ICMP router discovery in the Netsh routing IP routerdiscovery context. ICMP router discovery is provided by the Windows Server 2003 Routing and Remote Access service.

Commands in the Netsh Routing IP Routerdiscovery Context

Command Task

Add

Add interface. On the specified interface, specifies the following:

  • Whether ICMP router discovery is enabled or disabled.

  • A minimum and a maximum interval in minutes between router discovery packets.

  • The time in minutes after which a host considers the interface to be failed (which is after it has received its last Router Advertisement message from this router).

  • The preference level for using this interface as the default gateway. A higher number indicates a higher level of preference.

Delete

Delete interface. Disables ICMP router discovery configuration for the specified interface.

Dump

Creates a script that contains the current configuration of the ICMP router discovery component. If you save the script to a file, you can use it to restore altered configuration information.

If you plan to make extensive changes to the configuration of the ICMP router discovery component, a recommended practice is to begin the configuration session by using the dump command before you start, in case you need to restore the configuration later

Set

Set interface. On the specified interface, updates the following:

  • Whether ICMP router discovery is enabled or disabled.

  • A minimum and a maximum interval in minutes between router discovery packets.

  • The time in minutes after which a host considers the interface to be failed (which is after it has received its last Router Advertisement message from this router).

  • The preference level for using this interface as the default gateway. A higher number indicates a higher level of preference.

Show

Show interface. For the specified interface, displays:

  • Whether ICMP router discovery is enabled or disabled.

  • The minimum and maximum advertisement intervals.

  • The preference level for using this interface as the default gateway.

Uninstall

Removes ICMP router discovery from all interfaces on this server. If, later, you want to reenable ICMP router discovery by using the add interface command, you must add ICMP router discovery one interface at a time.

Management and Diagnostic Command-Line Tools for Unicast IPv4 Routing

In addition to the Netsh command-line tool, Windows Server 2003 also supports the Pathping, Ping, Route, and Tracert command-line tools for unicast IPv4 and IPv6 network traffic. This section describes each tool as used for unicast IPv4 routing–related tasks in an IPv4 internetworking environment. For information about these tools in an IPv6 internetworking environment, see IPv6 Tools and Settings.

Pathping.exe

Category The Pathping.exe diagnostic tool for IPv4 internetworks is included with the Windows Server 2003 operating system. The Windows Server 2003 version of Pathping.exe also supports IPv6.

Version compatibility The Pathping.exe tool runs on Windows Server 2003, Windows XP, and Windows 2000.

Pathping combines information provided by Tracert (identifying which routers exist along the path) and Ping (repeatedly sending messages to each router) with additional information that neither of those tools provides.

The Pathping tool uses ICMP Echo messages to test router latency, link latency, and packet losses on an IPv4 internetwork. Pathping tests each hop by pinging multiple times over a period of time, and then displaying delay and packet loss information per hop. Specifically, Pathping records the following:

  • The average round-trip time.

  • The packet loss when sending ICMP Echo messages to each router.

  • The packet loss when sending ICMP Echo messages across the links between each router.

The output of the Pathping command can help you determine if there is a high-loss link or router in the path by showing the degree of packet loss at each router or link.

For more information about Pathping.exe, see Command-Line Reference for Windows Server 2003, Standard Edition.

Ping.exe

Category The Ping.exe diagnostic tool for IPv4 internetworks is included with the Windows Server 2003 operating system. The Windows Server 2003 version of Ping.exe also supports IPv6.

Version compatibility The Ping.exe tool runs on Windows Server 2003, Windows XP, Windows 2000, and Windows NT® 4.0.

The Ping tool tests reachability on an IPv4 internetwork by sending an ICMP Echo message from a source host to a destination IP address and recording the results. Ping indicates whether the destination responds by displaying an ICMP Echo Response message and how much time elapses (in milliseconds) before a reply is received. If Ping receives no response, it displays an error message instead.

In addition to testing connectivity to a destination computer, successfully pinging from the local computer by IP address, not by name, verifies that TCP/IP is functioning properly on the local computer. Likewise, successfully pinging the local router verifies that the router is functioning.

For more information about Ping.exe, see Command-Line Reference for Windows Server 2003, Standard Edition.

Route.exe

Category The Route.exe management tool for IPv4 internetworks is included with the Windows Server 2003 operating system. The Windows Server 2003 version of Route.exe also supports IPv6.

Version compatibility The Route.exe tool runs on Windows Server 2003, Windows XP, Windows 2000, and Windows NT 4.0.

The Route tool displays and modifies entries in the IPv4 routing table on the local computer. You can use Route to add a temporary or persistent route, to change an existing route, or to remove a route from the IP routing table. The Route tool has four subcommands, listed in the following table.

Subcommands for Route.exe

Subcommand Task

Route print

Prints the IPv4 routing table and (if IPv6 exists) the IPv6 routing table.

You can also use route print to display only part of the routing table. For example, if you use route print to display the entire routing table and then decide that you want to look only at the set of routes that begin with, say, 157, you can use route print 157* to print only those routes that match 157*.

Route add

Adds a static route.

In Windows Server 2003, this subcommand does not support IPv6.

Route delete

Deletes a static route.

In Windows Server 2003, this subcommand does not support IPv6.

Route change

Modifies an existing static route. This command is used only to modify gateway or metric values.

In Windows Server 2003, this subcommand does not support IPv6

The IP routing table is stored in RAM, which means that a non-persistent static route disappears whenever a computer restarts. Typically, therefore, you use route add with the -p option to add a persistent static route that continues to exist even after a computer restarts.

Persistent routes are stored in the registry in the PersistentRoutes entry under HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters. Because you can add, modify, or delete a persistent static route by using the Route tool, you do not need to configure this registry entry directly.

For more information about Route.exe, see Command-Line Reference for Windows Server 2003, Standard Edition.

Tracert.exe

Category The Tracert.exe diagnostic tool for IPv4 internetworks is included with the Windows Server 2003 operating system. The Windows Server 2003 version of Tracert.exe also supports IPv6.

Version compatibility The Tracert.exe tool runs on Windows Server 2003, Windows XP, Windows 2000, and Windows NT 4.0.

The Tracert route tracking tool tests reachability, network latency, routing loops, and other issues on an IPv4 internetwork by determining the path (the series of routers) that unicast IPv4 traffic takes between a source host and destination IP address. Tracert sends ICMP Echo messages with incrementally higher values of the Time to Live (TTL) field in the IP header and then returns (lists) the series of near-side router interfaces on each router between the source and destination hosts. If the intended destination host does not receive the packets sent, Tracert displays the last router that successfully forwarded the packets.

For more information about Tracert.exe, see Command-Line Reference for Windows Server 2003, Standard Edition.

Network Connections

For unicast IPv4 routing, Network Connections is used to configure default gateway options and other options for TCP/IP.

Category Network Connections is included with the Windows Server 2003 operating system.

Version compatibility Network Connections runs on Windows Server 2003 and Windows XP. Network Connections is known as Network and Dial-up Connection in Windows 2000. Network Connections is known as the Network component of Control Panel in Windows NT 4.0.

The Network Connections management tool enables the configuration of TCP/IP properties for a variety of purposes. The following table describes only the options used for manually configuring the default gateway and the related automatic metric option.

TCP/IP Properties Options in Network Connections

Page Tab Task

TCP/IP Properties

General

Manually configuring a network adapter to point to a default gateway:

  • Under Use the following IP address, specify:

    • IP address and Subnet mask. Together, the IP address and subnet mask identify this computer and the subnet on which it is located.

    • Default gateway. The IP address for the default gateway identifies a local IP router (on the same subnet as this computer) that forwards traffic for this computer to destinations on a different subnet.

Advanced TCP/IP Settings

IP Settings

Adding, editing, or removing additional default gateways for this network connection:

  • Under Default gateways, click Add:

    Opens the TCP/IP Gateway Address dialog box, which you can use to configure an IP address and an automatic or manually configured metric. If the default gateway specified on the General tab fails, the local computer can switch to an alternative default gateway listed here if dead gateway detection is enabled.

    See the registry entry EnableDeadGWDetect later in this document.

Enabling the option to automatically calculate the interface cost based on the speed of the interface:

  • Select Automatic metric. An interface metric is enabled that is calculated automatically by TCP/IP based on the speed of the interface. If multiple closest matching routes to a destination exist in the IP routing table, IP uses the route with the lowest cost so that the fastest interface is used to forward traffic to the default gateway.

Disabling the option to automatically calculate the interface cost based on the speed of the interface:

  • Clear Automatic metric, and then specify an Interface metric. If you disable Automatic metric, the Interface metric field becomes available to manually configure a cost for this interface.

For related information about tools for configuring IP routers and hosts to enable hosts to dynamically discover the best default gateway to use on a subnet, see “Netsh routing IP routerdiscovery context” earlier in this document and see “ICMP router discovery options” later in this document.

Rrasmgmt.msc: Routing and Remote Access Snap-in

For unicast IPv4 routing, the Routing and Remote Access snap-in is used to configure options for a variety of functions provided by the Routing and Remote Access service. These functions include event logging, route source preference levels, the RIP for IP and OSPF routing protocols, the DHCP relay agent, IP packet filtering, and ICMP router discovery.

Category The Routing and Remote Access snap-in of Microsoft Management Console (MMC) is included with Windows Server 2003 operating system. Routing and Remote Access is disabled by default.

Version compatibility The Routing and Remote Access snap-in is provided by the Routing and Remote Access service on computers running Windows Server 2003 and Windows 2000 Server. For Windows NT 4.0, the Routing and Remote Access Service (RRAS) and its snap-in are available as a separate download from the Microsoft Windows NT Server Routing and Remote Access Service Download page.

You can use the Routing and Remote Access snap-in to enable, configure, or modify, unicast IPv4 routing options for RIP, OSPF, DHCP relay agent, IP packet filtering, and ICMP router discovery. The Routing and Remote Access snap-in can also be used to configure options for a variety of other routing or remote access services unrelated to these unicast IPv4 routing topics, such as configuring remote access connections, multicast IPv4 routing options, or network address translation (NAT).

Options for enabling, disabling, or refreshing the Routing and Remote Access service

The Routing and Remote Access snap-in is available by default in Administrative Tools in Control Panel. However, the Routing and Remote Access service managed by means of this snap-in is disabled by default.

Note

You do not use the Add or Remove Programs icon in Control Panel to install or uninstall the Routing and Remote Access service.

The following table describes the options to enable, disable, or refresh the Windows Server 2003 Routing and Remote Access service. These settings are available in the console tree of the Routing and Remote Access snap-in by right-clicking the ServerName node.

Options to Enable, Disable, or Refresh the Routing and Remote Access Service

Node Task

ServerName

Enabling the Routing and Remote Access service:

  • Configure and Enable Routing and Remote Access. Starts Routing and Remote Access Server Setup Wizard. To enable LAN routing, select Custom configuration on the wizard’s Configuration page, and then select LAN routing on the Custom configuration page.

    noteNote
    Alternatively, you can use the wizard to configure and enable connections for remote access dial-up or VPN, NAT, or a site-to-site VPN connection between two private networks from the Configuration page. If you select the Custom configuration option from the Configuration page, you can enable any of these connections in addition to LAN routing, or you can enable LAN routing only.

Disabling the Routing and Remote Access service:

  • Disable Routing and Remote Access. Disables the Routing and Remote Access service, which removes all Routing and Remote Access registry settings, but it does not uninstall the service or snap-in.

Refreshing the Routing and Remote Access service:

  • Disable Routing and Remote Access, and then Configure and Enable Routing and Remote Access. Refreshes the Routing and Remote Access service by first disabling it, which removes all current configuration, and then reenabling it, which starts the wizard so that you can re-enable the service and configure new settings. You do not need to restart the server.

General options

Options available in the IP Routing\General container in the Routing and Remote Access snap-in take effect at the level of the server rather than at the level of an individual component of the Routing and Remote Access service. General options include those for setting global logging levels and for ranking route sources. You can also set logging levels at the level of the individual component.

This section includes unicast IPv4 routing options available in the console tree on the IP Routing\General node of the Routing and Remote Access snap-in and in the details pane on the InterfaceName Properties page. Multicast routing options are not included in this section. For more information, see IPv4 Multicasting Tools and Settings.

General options in the console tree

The following table describes the options for adding a new unicast IPv4 routing protocol to the Routing and Remote Access snap-in console tree and for adding an interface to the details pane. These options are available by right-clicking the IP Routing\General node.

Options on the IP Routing\General Node in the Console Tree

Node Task

General

Adding a new routing protocol to the console tree:

  • New Routing Protocol. Provides a list from which to select one or more of the following protocols to add to the Routing and Remote Access console tree under the IP Routing node:

    • DHCP Relay Agent

    • IGMP Router and Proxy

    • RIP

    • OSPF

    • NAT/Basic Firewall

    This document describes the management of the DHCP relay agent, RIP, and OSPF components in the Routing and Remote Access snap-in. For information about IGMP, see IPv4 Multicasting Tools and Settings. For information about NAT/Basic Firewall, see NAT Tools and Settings.

Adding a new interface to the details pane:

  • New Interface. Provides a list of available interfaces on this server that you can add to the details pane of the IP Routing\General node.

The following table describes Routing and Remote Access options for configuring global IP logging and for ranking route source preference levels. These settings are available in the console tree of the Routing and Remote Access snap-in on the properties page of the IP Routing\General node. The properties page on the General node also includes a tab for Multicast Scopes, which is not covered in this technical reference.

Options on the IP Routing\General Properties Page

Tab Task

Logging

Configuring events to be logged in the System Log in Event Viewer for this server:

  • Log errors only

  • Log errors and warnings

  • Log the maximum amount of information (that is, errors, warnings, and informational messages)

  • Disable event logging

noteNote
Use information about routing events stored in the System Log to troubleshoot routing processes. Logging puts a considerable load on system resources and should be enabled only when needed. After the event is logged or the problem is identified, immediately reset logging to its default value, Log errors only.

Preference Levels

Configuring the rank of route sources in order of preference by moving one or more of the listed route sources up or down (route sources are listed here in default order):

  • Local

  • Static (a static route assigned to a demand-dial interface)

  • Static (non demand-dial) (a static route assigned to a LAN interface)

  • Auto-static

  • Network management (such as SNMP)

  • Open Shortest Path First (OSPF)

  • RIP Version 2 for Internet Protocol

Note

If two routes exist to the same destination, the route learned by the route source with a higher preference is preferred regardless of the route metric. For example, if the preference level for OSPF is higher than the preference level for RIP v2, an OSPF route with a metric of 5 is entered into the routing table rather than the same route discovered by RIP with a metric of 3.

General options on the InterfaceName Properties page

For a description of unicast IPv4 routing configuration options available on the General tab of the InterfaceName Properties page in the IP Routing\General container, see the separate sections for “IP Packet Filtering Options” and “ICMP router discovery options” later in this document.

This InterfaceName Properties page also provides the Configuration tab to specify whether the IP address for this interface is statically configured or obtained dynamically from a DHCP server, and it provides the Multicast Boundaries and Multicast Heartbeat tabs to configure options related to IPv4 multicast routing. None of these options is described here because they are outside the scope of this technical reference.

RIP Options

Routing Information Protocol (RIP), the primary dynamic routing protocol used on small or medium-sized IPv4 internetworks (10 to 50 networks), enables the dynamic exchange of IPv4 routing information between RIP routers. By adding RIP to the Routing and Remote Access snap-in console tree, adding RIP interfaces to the details pane, and then configuring RIP options, you can enable RIP routers to update topology changes as networks are added or removed and as links fail and are restored.

The Windows Server 2003 Routing and Remote Access service supports both the updated version of RIP, RIP version 2 (RIP v2) and the earlier version of RIP, RIP version 1 (RIP v1). The version of RIP installed when you add the RIP protocol to the Routing and Remote Access snap-in console tree is RIP v2, which includes options that make RIP v1 optionally available.

This section includes unicast IPv4 routing options available in the console tree of the Routing and Remote Access snap-in on the IP Routing\RIP node and in the details pane on the InterfaceName Properties page.

RIP options in the console tree

The following table describes options for adding one or more interfaces to the details pane and for displaying information about neighboring RIP routers. These options are available by right-clicking the IP Routing\RIP node.

Options on the IP Routing\RIP Node in the Console Tree

Node Task

RIP

Adding a new interface to the details pane:

  • New Interface. Provides a list of available interfaces on this server that you can add to the details pane of the IP Routing\RIP node.

Displaying neighboring RIP routers:

  • Show Neighbors. Displays a table of RIP routers configured as neighbor routers to this RIP router. For each RIP neighbor, the table lists its IP address, RIP version, and any bad packets or bad routes.

The following table describes Routing and Remote Access options for the RIP routing protocol. These settings are available in console tree of the Routing and Remote Access snap-in on properties page of the IP Routing\RIP node.

Options on the RIP Properties Page

Tab Task

General

Configuring the delay before this RIP router sends triggered updates:

  • Maximum delay (seconds). Specifies the maximum amount of time that this router waits before it sends triggered updates (updates that are announced as soon as a network topology change occurs, rather than at periodic intervals). The default value is 5 seconds.

Configuring events to be logged in the System Log in Event Viewer for this RIP router:

  • Log errors only

  • Log errors and warnings

  • Log the maximum amount of information (that is, errors, warnings, and informational messages)

  • Disable event logging

noteNote
Use information about routing events stored in the System Log to troubleshoot routing processes. Logging puts a considerable load on system resources and should be enabled only when needed. After the event is logged or the problem is identified, immediately reset logging to its default value, Log errors only.

Security

Configuring the RIP routers from which this RIP router accepts announcements:

  • Accept announcements from all routers (default). Specifies that this RIP router accepts announcements from all RIP routers.

  • Accept announcements from listed routers only. Specifies that this RIP router accepts only announcements from the listed RIP routers.

  • Ignore announcements from all listed routers. Specifies that this RIP router ignores announcements from the listed RIP routers but accepts announcements from all other RIP routers.

  • Router IP address. Specifies the list (if any) of routers whose announcements you want to accept or ignore.

RIP for IP options on the InterfaceName Properties page

The following table describes Routing and Remote Access options for a RIP interface. These settings are available in the Routing and Remote Access snap-in on the specified tab of the InterfaceName Properties page for an interface in the IP Routing\RIP container.

RIP Options on the InterfaceName Properties Page

Tab Task

General

Configuring Operation mode options for this interface:

  • Auto-static update mode (default for demand-dial interfaces). Sends RIP announcements only when other routers request an update. Routes learned are marked as static routes in the IP routing table and remain in the routing table until they are manually deleted.

  • Periodic update mode (default for LAN interfaces). Sends RIP announcements periodically every n seconds as specified in Periodic announcement interval on the Advanced tab. Routes learned are marked as RIP routes in the IP routing table and are removed if the router restarts.

Configuring Outgoing packet protocol options for sending outgoing RIP route announcements over this interface:

  • RIP version 1 broadcast. Broadcasts RIP v1 announcements. Specify this protocol only if all neighboring RIP routers are also RIP v1 routers.

  • RIP version 2 broadcast. Broadcasts RIP v2 announcements. Specify this protocol for a mixed environment that uses both RIP v1 and RIP v2.

  • RIP version 2 multicast. Multicasts RIP v2 announcements. Specify this protocol only if all neighboring RIP routers are also RIP v2 routers.

  • Silent RIP. Disables outgoing RIP announcements on this interface. Specify this option to configure this non-router computer to listen for incoming RIP announcements and update its routing table but not to announce its own routes.

Configuring Incoming packet protocol options for receiving incoming RIP announcements over this interface:

  • Ignore incoming packets. Configures this router only to announce routes (but not to receive incoming route announcements).

  • RIP version 1 and 2. Configures this router to accept both RIP v1 and RIP v2 announcements.

  • RIP version 1 only. Configures this router to accept only RIP v1 announcements.

  • RIP version 2 only. Configures this router to accept only RIP v2 announcements.

Configuring metric and route tag options for this interface:

  • Added cost for routes. Configures the cost associated with sending packets over this interface. For example, set a higher cost on a heavily trafficked route to reduce its use. Increasing the cost for all RIP routes reduces your maximum RIP network diameter.

  • Tag for announced routes. Specifies a tag number for the routes advertised on this interface in order to distinguish these routes from non-RIP routes. Used on a router that can support multiple routing protocols.

Configuring the simple password authentication option for this interface:

  • Activate authentication. Enables a form of authentication that uses a plaintext password for RIP v2 announcements over this interface. All incoming and outgoing RIP v2 packets from neighboring RIP v2 routers connected to this interface must contain the same plaintext password. This type of authentication uses plaintext passwords, which are used by routers to identify each other, not for security.

  • Password. Specifies a plaintext password (1–16 characters).

Security

Configuring RIP security options for incoming routes on this interface:

  • Accept all routes. Configures this RIP router to look at each route entry in an incoming RIP announcement to determine whether to update its routing table.

  • Accept all routes in the ranges listed. Configures this RIP router to look at each route entry in an incoming RIP announcement but processes the route only if it falls within one of the ranges listed.

  • Ignore all routes in the ranges listed. Configures this RIP router to look at each route entry in an incoming RIP announcement and discards the route if it falls into one of the ranges listed.

Configuring RIP security options for outgoing routes on this interface:

  • Accept all routes. Configures this RIP router to include all appropriate route entries in outgoing RIP announcements.

  • Accept all routes in the ranges listed. Configures this RIP router to include a route in the outgoing RIP announcement only if the route falls into one of the ranges listed.

  • Ignore all routes in the ranges listed. Configures this RIP router to exclude a route from the outgoing RIP announcement if the route falls into one of the ranges listed.

Neighbors

Configuring how this RIP router interacts with neighboring RIP routers through this interface:

  • Use broadcast or multicast only. Specifies that all RIP announcements are broadcasted or multicasted as specified in Outgoing packet protocol on the General tab.

  • Use neighbors in addition to broadcast or multicast. Specifies that RIP announcements are unicasted to specific neighboring routers in addition to broadcasted or multicasted as specified in Outgoing packet protocol on the General tab.

  • Use neighbors instead of broadcast or multicast. Specifies that RIP announcements are unicasted only to specific neighboring routers. For nonbroadcast networks, such as Frame Relay, select this option to enable RIP traffic between specific routers.

  • IP address. Specifies the IP address of one or more neighboring RIP routers. Valid only if one of the preceding options to send RIP announcements only to specific neighboring routers is selected.

Advanced

Configuring RIP announcement and route interval options for this interface:

  • Periodic announcement interval (seconds). Specifies the interval — from 15 seconds through 86,400 seconds (24 hours) — between periodic RIP announcements. The default value is 30 seconds. Valid only if Periodic update mode is selected as the Operation mode on the General tab.

  • Time before routes expire (seconds). Specifies the lifetime — from 15 seconds through 259,200 seconds (72 hours) — of a route that is learned through RIP. The default value is 180 seconds. If a route is not updated with another RIP announcement within the specified time, the route expires. Valid only if Periodic update mode is selected as the Operation mode on the General tab.

  • Time before route is removed (seconds). Specifies the amount of time — from 15 seconds through 259,200 seconds (72 hours) — after which a RIP-learned route that has expired is removed from the routing table. The default value is 120 seconds. Valid only if Periodic update mode is selected as the Operation mode on the General tab.

Configuring split-horizon processing and triggered updates options for this interface:

  • Enable split-horizon processing. Enabled by default. Specifies that routes learned from a neighboring RIP router are not announced in RIP announcements that this router sends to that neighbor (unless poison-reverse processing is also enabled, in which case such routes are announced with a metric of 16 to indicate explicitly that those routes are unavailable). Helps reduce convergence time and avoid routing loops.

    • Enable poison-reverse processing. Enabled by default. Specifies that routes learned from a neighboring RIP router are announced with a metric of 16 (unreachable) in RIP announcements that this router sends to that neighbor. Available only if split horizon is enabled (if you disable split-horizon, poison-reverse processing is also disabled).

  • Enable triggered updates. Specifies that a change in network topology, such as a new route or a change in metric, triggers an immediate update announcement that includes only the change. Configure Maximum delay (seconds) between triggered updates on the General tab on the RIP Properties page.

  • Send clean-up updates when stopping. Enabled by default. Specifies that, if this RIP router stops, the router advertises as unreachable all of the routes that it makes available on all of its interfaces as it shuts down.

Configuring host and default route options for this interface:

  • Process host routes in received announcements. Specifies that host routes in received RIP announcements are accepted. By default, host routes are ignored.

  • Include host routes in sent announcements. Specifies that host routes are included in RIP announcements. By default, host routes are not included.

  • Process default routes in received announcements. Specifies that default routes in received RIP announcements are accepted. By default, default routes are ignored.

  • Include default routes in sent announcements. Specifies that default routes are included in RIP announcements. By default, default routes are not included.

Configuring subnet summarization option for this interface:

  • Disable subnet summarization. Selected by default. Specifies that subnet routes are not summarized in the form of the class-based network ID when announced on a network that is not a subnet of the class-based network ID. Available only if Outgoing packet protocol on the General tab is set to either RIP version 2 broadcast or RIP version 2 multicast. In a mixed environment that contains both RIP v1 and RIP v2 routers, use only contiguous, fixed-size subnets, or clear this option.

OSPF options

Open Shortest Path First (OSPF), the most common routing protocol used on large and very large IPv4 internetworks, enables the dynamic exchange of IPv4 routing information between OSPF routers across complex internetworks. Adding OSPF to the Routing and Remote Access snap-in console tree, adding OSPF interfaces to the details pane, and then configuring OSPF options enables OSPF routers to update topology changes as networks are added or removed and as links fail and are restored.

This section includes unicast IPv4 routing options available in the console tree of the Routing and Remote Access snap-in on the IP Routing\OSPF node and in the details pane on the InterfaceName Properties page.

OSPF options in the console tree

The following table describes options for adding one or more interfaces to the details pane and for displaying information about the OSPF autonomous system. These options are available by right-clicking the IP Routing\OSPF node.

Options on the IP Routing\OSPF Node in the Console Tree

Node Task

OSPF

Adding a new interface to the details pane:

  • New Interface. Provides a list of available interfaces on this server that you can add to the details pane of the IP Routing\OSPF node.

Displaying information about the OSPF autonomous system:

  • Show Areas. Displays a table of OSPF areas, including:

    • Area ID. The identification number of each OSPF area.

    • State. Up or down.

    • SPF computations. The number of Shortest Path First (SPF) computations. An OSPF router creates a tree (the SPF tree) of shortest paths to each other router and subnet in the OSPF autonomous system.

  • Show Link-state Database. Displays a table of information about the OSPF LSDB, including:

    • Area ID. The identification number of the OSPF area for each link state advertisement (LSA) in the table.

    • Type. The type of advertisement, such as Router, Summary (Network), Stub, Network, Summary (ASBR), or AS External.

    • Link state ID. The identification number of each advertisement in the table. The LSAs vary by advertisement type.

    • Advertising router. The OSPF router ID of the advertising router

    • Age. The age (in seconds) since the advertisement originated.

    • Sequence. A number that identifies old or duplicate advertisements.

  • Show Neighbors. Lists information about neighboring OSPF routers, including:

    • Neighbor. The IP address of each neighboring router.

    • Neighbor ID. Each neighboring router’s OSPF router ID.

    • Type. The network type of each neighboring router, such as NBMA.

    • State. The state of each neighbor router:Down, Attempt, Init, 2-way, ExStart, Exchange, Loading, or Full. For a definition of each of these states, see “How OSPF Works” in How Unicast IPv4 Routing Protocols and Services Work.

    • Priority. The router priority of each neighboring router.

    • State-changes and errors. The total number of changes of state and errors.

    • Retransmit-queue length. The length of the retransmit queue.

  • Show Virtual Interfaces. Displays a table of virtual interfaces (virtual links). A virtual link is a logical connection that enables the exchange of routing information between a backbone area border router (ABR) and an ABR that is not physically connected to the backbone area across a transit area. You cannot configure a virtual link across the backbone area. For each virtual link, the table includes:

    • Area ID. The area ID of the transit area for the virtual link.

    • Neighbor ID. The neighboring router’s OSPF router ID.

    • State. The state of each virtual interface:Down, Loopback, Waiting, Point-to-point, Designated-router, Backup designated-router, Other, or Unknown.

    • Sent packets. The number of packets sent over this virtual link.

    • Received packets. The number of packets received over this virtual link.

    • Dropped packets. The number of packets that failed to be transmitted over this virtual link.

The following table describes Routing and Remote Access options for OSPF properties. These settings are available in the console tree of the Routing and Remote Access snap-in on the properties page of the IP Routing\OSPF node.

Options on the OSPF Properties Page

Tab Task

General

Configuring the OSPF router ID:

  • Router identification. Specifies the 32-bit number (in dotted decimal notation) that uniquely identifies this OSPF router on the network. The OSPF router ID is not an IP address. Ensure that the number is unique, for example, by selecting the highest or the lowest IP address assigned to one of the interfaces on this computer.

Configuring this OSPF router as an autonomous system boundary router (ASBR):

  • Enable autonomous system boundary router. Specifies that the OSPF router is an ASBR and therefore advertises routing information into the OSPF autonomous system that is from route sources external to the OSPF autonomous system. If you select this option, you can also use the External Routing tab to specify the route sources — such as static routes or the Routing Information Protocol (RIP) — from which this router will accept routes.

Configuring events to be logged in System Log in Event Viewer for this OSPF router:

  • Log errors only

  • Log errors and warnings

  • Log the maximum amount of information (that is, errors, warnings, and informational messages)

  • Disable event logging

noteNote
Use information about routing events stored in the System Log to troubleshoot routing processes. Logging puts a considerable load on system resources and should be enabled only when needed. After the event is logged or the problem is identified, immediately reset logging to its default value, Log errors only.

Areas

Configuring OSPF areas (each OSPF interface on this OSPF router must belong to at least one OSPF area):

  • Add. Opens the OSPF Area Configuration dialog box, which you can use to configure the following options:

    On the General tab:

    • Area ID. Specifies an ID (a 32-bit number expressed in dotted decimal notation) that identifies the OSPF area. The only required OSPF area is the OSPF backbone area, which requires the reserved area ID of 0.0.0.0. An OSPF backbone area is either the sole area in the OSPF autonomous system, or it is the central area to which all other OSPF areas in this OSPF autonomous system connect and through which packets are routed from one non-backbone area to another non-backbone area.

      Although an OSPF area ID is neither an IP address nor an IP network ID, if this is not the backbone area and if this area can be summarized with a single network ID, you can set the area ID to match the single network ID. For example, if an area contains all of the subnets for the IP network 10.1.0.0/16, you can set the area ID to 10.1.0.0.

    • Enable plaintext password. Enabled by default. Specifies that a plaintext password is required for the OSPF area so that OSPF routers can identify each other (this is not a security option). If this password option is enabled, all interfaces in this area that are on the same subnet must use the same password. By default, the password is 12345678.

    • Stub area. Specifies that this area is a stub area and therefore does not accept external routes from outside this OSPF autonomous system. Routing from a stub area to destinations outside this OSPF area, which includes destinations in other areas within its own autonomous system as well as to destinations outside its autonomous system, uses a summary default route. You cannot configure the backbone as a stub area, nor can you configure a virtual link through a stub area.

    • Stub metric. Specifies the cost of the summary default route that this OSPF router uses to reach all destinations external to this OSPF autonomous system.

    • Import summary advertisements. Specifies that inter-area routes for the OSPF autonomous system are imported into this stub area. By default, this check box is cleared and all non-intra-area destinations within the OSPF autonomous system are routed based on a single default route.

    On the Ranges tab:

    • Destination and Network mask. Specifies the network ID and subnet mask pair (that is, a summary route) that defines a range of addresses that belongs to this OSPF area. OSPF uses OSPF ranges to summarize the routes within this OSPF area.

  • Edit. Opens the OSPF Area Configuration dialog box, which you can use to modify settings for the selected OSPF area. These settings are the same as those listed just above for Add on the General tab and on the Ranges tab.

  • Remove. Removes the selected OSPF area.

Virtual Interfaces

Configuring a virtual interface for this OSPF router:

Note

A virtual interface, also called a virtual link, is a logical connection that enables the exchange of routing information between a backbone area border router (ABR) and an ABR that is not physically connected to the backbone area. You cannot configure a virtual link across the backbone area or a stub area.

  • Add. Opens the OSPF Virtual Interface Configuration dialog box, which you can use to create a virtual link by configuring the following options:

    On the General tab:

    Note

    The following settings must be the same on both routers at each end of the virtual link.

    • Transit area ID. The OSPF area ID for the transit area for this virtual link. Select the one that you want to use from the provided list of non-backbone area IDs configured on this OSPF router.

    • Virtual neighbor router ID. The OSPF router ID of the router at the other end of the virtual link.

    • Transit delay (seconds). The estimated number of seconds that it takes to transmit OSPF packets to the virtual link neighbor router over this interface. The default is 1 second.

    • Retransmit interval (seconds). The interval that the OSPF router waits before retransmitting link state messages. This value should be substantially larger than the expected round-trip delay between the two routers at either end of the transit area. The default is 10 seconds.

    • Hello interval (seconds). The interval between hello packets that are sent out over this interface. The default is 10 seconds.

    • Dead interval (seconds). The interval that a router’s neighbors wait after they cease to hear the router’s hello packets before they declare that the router is down. The default is 60 seconds.

    • Plaintext password. The plaintext password to be shared by both routers at each end of the virtual link. This password is not the same as the plaintext password shared by all routers in an OSPF area so that OSPF routers in that area can identify each other.

  • Edit. Opens the OSPF Virtual Interface Configuration dialog box, which you can use to modify the selected virtual interface. These settings are the same as those listed for the Add option.

  • Remove. Removes the selected virtual interface.

External Routing

Configuring external route source options:

  • Accept routes from all route sources except those selected. Specifies that routes from all selected route sources are ignored.

  • Ignore routes from all route sources except those selected. Specifies that only routes from selected route sources are accepted.

  • Route sources. Specifies non-OSPF route sources from which this OPSF router accepts or ignores routes:

    • Auto-static routes

    • Local routes

    • RIP version 2 for Internet Protocol

    • SNMP routes

    • Static (non-demand dial) Routes (refers to a static route assigned to a LAN interface)

    • Static routes (refers to a static route assigned to a demand-dial interface)

  • Route filters. Opens the OSPF External Route Filters dialog box, which you can use to specify filters that accept or ignore listed external routes, and to specify the list of routes (by destination and subnet mask) to be accepted or ignored.

OSPF options on the InterfaceName Properties page

The following table describes Routing and Remote Access options for an OSPF interface. These settings are available in the Windows Server 2003 Routing and Remote Access snap-in on the specified tab of the InterfaceName Properties page for an interface in the IP Routing\OSPF container.

OSPF Options on the InterfaceName Properties Page

Tab Task

General

Configuring OSPF on this interface:

  • Enable OSPF for this address. Displays the IP address assigned to this interface, and specifies whether OSPF is enabled on this interface. If two or more IP addresses are assigned to the interface (by using the Advanced TCP/IP Settings page in TCP/IP Properties), select the IP address that you want to configure from the list provided. If more than one IP address is configured for this interface, you must configure OSPF settings on the General, NBMA Neighbors, and Advanced tabs for each IP address. Specify the following:

    • Area ID. The OSPF area ID for the interface. Select the one that you want to use from the provided list of available areas.

    • Router priority. The OSPF router priority for the interface. When two or more OSPF routers attached to a subnet that does not have an existing designated router attempt to become the designated router of the subnet, the router with the highest router priority setting takes precedence. If a tie exists, the router with the higher router ID takes precedence. An interface with a priority of 0 cannot become the designated router. The priority setting is ignored for point-to-point links.

    • Cost. The cost, advertised in the router’s link state advertisement (LSA), of sending a packet out over this interface. Because an interface with a lower cost is used more frequently, typically, you configure a faster interface with a lower cost.

    • Password. The plaintext password for this OSPF area, which is enabled by default. All interfaces in the same OSPF area that are on the same subnet must use the same password.

Configuring the network type on this interface by selecting one of the following:

  • Broadcast. Specifies that the interface is a broadcast OSPF interface on a multiple access broadcast network in which multiple routers can be connected to the network. Ethernet, Token Ring, and FDDI are examples of broadcast networks.

  • Point-to-point. Specifies that this interface is a point-to-point OSPF interface that connects a single pair of routers. T1/E1, T3/E3, ISDN, and dial-up links are examples of point-to-point networks.

  • Non-broadcast multiple access (NBMA). Specifies that this interface is an NBMA OSPF interface. If you specify this type of interface, you can then configure neighboring routers by using the NBMA Neighbors tab. X.25, Frame Relay, and Asynchronous Transfer Mode (ATM) are examples of NBMA networks.

NBMA Neighbors

Configuring NBMA neighbors (available if Non-broadcast multiple access (NBMA) is selected as the network type on this interface’s General tab) for this interface:

  • IP address. Displays the IP address assigned to the interface. If two or more IP addresses are assigned to the interface, select the IP address that you want to configure from the list provided. If multiple IP addresses are configured for the interface, you must configure settings for each IP address.

  • Neighboring IP address. Specifies the IP address (not the OSPF router ID) of a neighbor router on the NBMA network that you want to add.

  • Router priority. Specifies the OSPF router priority of the neighbor router. A neighbor is eligible to become the designated router if its router priority is 1 or greater.

Advanced

Configuring advanced OSPF options for this interface:

  • IP address. Displays the IP address assigned to the interface. If two or more IP addresses are assigned to the interface, select the IP address that you want to configure from the list provided. If multiple IP addresses are configured for the interface, you must configure settings for each IP address.

  • Transmit delay (seconds). The estimated number of seconds it takes for this OSPF router to transmit an OSPF link state update packet over this interface. This value should take into account the transmission and propagation delays of the interface and network media. The default setting is 1 second.

  • Retransmit interval (seconds). The number of seconds between link state advertisement (LSA) retransmissions on this interface. This value should exceed the expected round-trip delay between any two routers on the attached subnet; if this value is too low, needless retransmissions result. The value needs to be higher on low-speed serial lines. A sample value for a local area network (LAN) subnet is 5 seconds.

  • Hello interval (seconds). The interval in seconds between transmissions of OSPF Hello packets on this interface. This setting must be the same for all routers on the same subnet. The shorter the hello interval, the faster topological changes are detected. However, a shorter interval also results in more OSPF traffic. A sample value for an X.25 network is 30 seconds. A sample value for a LAN is 10 seconds.

  • Dead interval (seconds). The number of seconds before a neighbor router determines that this router is unavailable. The router is determined to be unavailable if a neighboring router does not receive a Hello packet sent by this router within the specified interval. For example, if the hello interval setting is 15 seconds and the dead interval is 60 seconds, after four unreceived Hello packets, neighbor routers declare that this router is unavailable. Typically, the value for this setting is four times the value in the hello interval. This value must be the same for all OSPF router interfaces attached to the same subnet.

  • Poll interval (seconds). For NBMA interfaces only, the number of seconds between OSPF network polls. If a neighboring router is unavailable, it might still be necessary to send Hello packets to the unavailable neighbor in order to determine if the connection has been restored. This is called the poll interval. Typically, the poll interval is at least twice as long as the dead interval. A sample value for a poll interval on an X.25 network is 2 minutes.

  • Maximum transmission unit (MTU) size (bytes). The maximum size of an IP datagram carrying OSPF information that can be sent without fragmentation. The default value of 1,500 bytes is the default IP MTU for an Ethernet network.

DHCP Relay Agent options

Any IPv4 subnet that contains DHCP clients requires either a DHCP server or a DHCP relay agent to provide address leases to those DHCP clients. A computer acting as a DHCP relay agent, also known as a Bootstrap Protocol (BOOTP) relay agent, relays DHCP messages between DHCP clients on an IP subnet that has no DHCP server and a DHCP server located on a different subnet. By adding the DHCP relay agent to the Routing and Remote Access snap-in console tree, adding one or more DHCP relay agent interfaces to the details pane, and then configuring DHCP relay agent options, you can enable a router to act as a DCHP relay agent. Installation of a DHCP relay agent makes it unnecessary to install a separate DHCP server on each subnet in an IP internetwork.

This section includes unicast IPv4 routing options available in the console tree in the Routing and Remote Access snap-in on the IP Routing\DHCP Relay Agent node and in the details pane on the InterfaceName Properties page.

DHCP relay agent options in the console tree

The following table describes the option for adding one or more interfaces to the details pane. This option is available by right-clicking the IP Routing\DHCP Relay Agent node.

Options on the IP Routing\DHCP Relay Agent Node in the Console Tree

Node Task

DHCP Relay Agent

Adding a new interface to the details pane:

  • New Interface. Provides a list of available interfaces on this server that you can add to the details pane of the IP Routing\DHCP Relay Agent node.

The following table describes Routing and Remote Access options for DHCP relay agent properties. These settings are available in the console tree of the Routing and Remote Access snap-in on the properties page of the IP Routing\DHCP Relay Agent node.

Options on the DHCP Relay Agent Properties Page

Tab Task

General

Configuring one or more DHCP servers for local clients to contact through the DHCP relay agent:

  • Server address. Adds the IP address of one or more DHCP servers to create a list of available DHCP servers on the network. This router acts as a DHCP relay agent by relaying DHCP messages between DHCP clients on the local subnet and the DHCP server or servers specified in this list.

DHCP relay agent options on the InterfaceName Properties page

The following table describes Routing and Remote Access options for a DHCP relay agent interface. These settings are available in the Windows Server 2003 Routing and Remote Access snap-in on the General tab of the InterfaceName Properties page for an interface in the IP Routing\DHCP Relay Agent container.

DHCP Relay Agent Options on the InterfaceName Properties Page

Tab Task

General

Configuring a DHCP relay agent interface over which to route packets exchanged between DHCP clients on the local subnet and a DHCP server on another subnet:

  • Relay DHCP packets. Enables DHCP messages to be relayed between DHCP clients on the subnet to which this interface is attached and a DHCP server on a different subnet. Enabled by default when you add an interface to the IP Routing\DHCP Relay Agent node in the console tree.

  • Hop-count threshold. Specifies the maximum number of DHCP relay agents that can handle DHCP relayed traffic. The default value is 4 hops; the maximum value is 16 hops.

  • Boot threshold (seconds). Specifies the number of seconds the DHCP relay agent waits before forwarding DHCP messages. The default value is 4 seconds. This option is useful when you want a local DHCP server to respond first to a DHCP client’s request for an IP address, but, if no local DHCP server responds, you want the DHCP relay agent to forward the message to a remote DHCP server.

IP packet filtering options

Configuring IP packet filters on an IP router helps increase security on a private IP internetwork that connects to a public network such as the Internet. When IP packet filtering is configured, a packet arriving at a router might be forwarded to its destination, rejected with an error message returned to notify the sender, or dropped with no error message sent.

The following table describes options for IP packet filtering available in the Windows Server 2003 Routing and Remote Access snap-in by selecting Inbound Filters or Outbound Filters on the General tab of the InterfaceName Properties page for an interface in the IP Routing\General container, and then selecting New or Edit. Note that, although the options for adding an inbound or an outbound filter shown in the table appear together in the same dialog box, typically filters are configured in pairs, either {Destination Network/Protocol}, or {Source Network/Protocol}. After you add a filter, you select one of the following filter actions:

  • Receive all packets except those that meet the criteria below.

  • Drop all packets except those that meet the criteria below.

If you configure multiple parameters on a particular filter, when the filter is applied to an incoming packet, the parameters of the filter are compared by using a logical AND operator. The fields in the packet must match all of the criteria of the filter to have the filter action applied.

IP Packet Filtering Options on the InterfaceName Properties Page

Tab Task

General

Adding inbound filters or outbound filters (in the Add IP Filter or Edit IP Filter dialog box):

  • Source Network field:

    • IP address. The source network ID or IP address.

    • Subnet mask. The subnet mask that corresponds to the source network ID, or 255.255.255.255 for a source IP address. The subnet mask bits must encompass all of the bits used in the IP address field. The IP address cannot be more specific than the subnet mask, that is, the IP address (in binary notation) cannot have a bit set to 1 when the corresponding bit in the subnet mask is set to 0.

  • Destination Network field:

    • IP address. The destination network ID or IP address.

    • Subnet mask. The subnet mask that corresponds to the destination network ID, or 255.255.255.255 for a destination IP address. The subnet mask bits must encompass all of the bits being used in the IP address field. The IP address cannot be more specific than the subnet mask, that is, the IP address (in binary notation) cannot have a bit set to 1 when the corresponding bit in the subnet mask is set to 0.

  • Protocol field:

    • TCP. Displays text boxes in which to specify a source TCP port and a destination TCP port. One or both can be specified. If nothing is specified in these text boxes, they default to 0, which signifies any port.

    • TCP [established]. Displays text boxes in which to specify a source TCP port and a destination TCP port. Use this option when you want to define TCP traffic (source TCP port and destination TCP port) for TCP connections already established by the router.

      TCP [established] is used, for example, in a site-to-site VPN or dial-up connection when a calling router connects to an answering router. If you specify TCP [established], TCP traffic is accepted only if the calling router initiates the TCP connection.

    • UDP. Displays text boxes in which to specify a source UDP port and a destination UDP port. One or both can be specified. If nothing is specified in these text boxes, they default to 0, which signifies any port.

    • ICMP. Displays text boxes in which to specify an ICMP code and an ICMP type. One or both can be specified. If nothing is specified in these text boxes, they default to 255, which signifies any code and any type.

    • Any. Configures a filter for any protocol (Any is the default).

    • Other. Lets you specify the IP protocol identifier number for the protocol you want (rather than select a protocol by name). For example, TCP is IP protocol 6, UDP is IP protocol 17, and ICMP is IP protocol 1.

noteNote
For a complete list of IP protocol numbers, which are used to configure firewalls, routers, and proxy servers, see the Internet Assigned Numbers Authority (IANA) list of <A HREF="https://go.microsoft.com/fwlink/?LinkID=3858" TARGET="_blank">IP protocol numbers</A> at https://go.microsoft.com/fwlink/?LinkID=3858. (The IPv4 “Protocol” field is called the “Next Header” field in IPv6.)

Configuring fragmentation checking:

  • Enable fragmentation checking. If selected, specifies that this router discards all fragmented IP packets that it receives on this interface. This option applies only to incoming traffic.

For examples of filter configurations for commonly implemented IP filtering scenarios on a Routing and Remote Access router, see “Filtering Scenarios” in How Unicast IPv4 Routing Protocols and Services Work.

ICMP router discovery options

Configuring ICMP router discovery, an alternative to using manually configured or DHCP-configured default gateways, provides an improved method of configuring and detecting the default gateway. For IP hosts, the default gateway is the IP router on the local subnet that forwards traffic to destinations that are not located on the local subnet.

The ICMP router discovery feature automates the discovery and configuration of the default gateway so that IP hosts can dynamically discover the best default gateway to use on a subnet and can automatically switch to another default gateway if the first default gateway fails or if the network administrator changes router preferences.

The following table describes Routing and Remote Access options for configuring ICMP router discovery on an IP router. These settings are available in the Windows Server 2003 Routing and Remote Access snap-in on the General tab of the InterfaceName Properties page for an interface in the IP Routing\General container.

ICMP Router Discovery Options on the InterfaceName Properties Page

Tab Task

General

Configuring options for ICMP router discovery on this interface:

  • Enable router discovery advertisements. Specifies whether ICMP router discovery advertisements are enabled on the interface. If selected, you must also specify settings for the following (or accept the defaults):

    • Advertisement lifetime (in minutes). The interval after which a host determines that a router is unavailable (since its last Router Advertisement message). The default is 30 minutes.

    • Level of preference. The preference level of this router to be the default gateway. The router with the highest preference level becomes the default gateway. The default is 0, which means that any other router configured with 1 or a higher number is preferred over this router by router discovery clients.

    • Advertisement interval minimum time (in minutes). The minimum amount of time between Router Advertisement messages sent by this router. The default is 7 minutes.

    • Advertisement interval maximum time (in minutes). The maximum amount of time between Router Advertisement messages sent by this router. The default is 10 minutes. Router Advertisements are sent at a random interval between the minimum and maximum times.

In addition to enabling ICMP router discovery on the IP router, you must also configure router discovery on IP client computers on that router’s subnet:

  • For information about the tool used to configure ICMP router discovery on a client running Windows XP Professional, see “Netsh routing IP routerdiscovery context” in the section “Command-Line Tools for Unicast IPv4 Routing” earlier in this document.

  • For information about the tool used to configure ICMP router discovery on a client running Windows 2000 Professional, see “PerformRouterDiscovery” and “SolicitationAddressBcast” in the section Unicast IPv4 Routing Registry Entries later in this document.

Unicast IPv4 Routing Registry Entries

When the Routing and Remote Access service is enabled, it creates and maintains its settings in the Windows Server 2003 registry. For performance reasons, most Routing and Remote Access service configuration information is stored in binary format in large configuration blocks, not as separate registry entries that you can easily view or modify by using the Registry Editor (Regedit.exe) tool. This section includes only unicast IPv4 routing-related Routing and Remote Access registry entries that are configurable.

For more information about these and other registry entries, see the Windows Server 2003 Resource Kit Registry Reference.

The information here is provided as a reference for use in troubleshooting or verifying that the required settings are applied. It is recommended that you do not directly edit the registry unless there is no other alternative. Modifications to the registry are not validated by the registry editor or by Windows before they are applied, and as a result, incorrect values can be stored. This can result in unrecoverable errors in the system. When possible, use Group Policy or other Windows tools, such as Microsoft Management Console (MMC), to change configuration settings rather than editing the registry directly. If you must edit the registry, use extreme caution.

HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Tracing

The Routing and Remote Access service provides an extensive tracing capability to help troubleshoot complex network problems. Separate routing components can be independently enabled to log tracing information to files — this capability is called file tracing. File tracing includes information about internal component variables, function calls, and interactions. You can enable and disable tracing for routing protocols while the router is running. This section includes only file tracing registry keys for the RIP v2 and OSPF routing protocols.

The following registry keys are located under the registry key HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Tracing.

IPRIP2

Registry path HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Tracing

Version Windows Server 2003 and Windows 2000 Server

Configuring entries under the IPRIP2 key enables file tracing, which logs tracing information about the RIP v2 routing protocol to files so that you can troubleshoot problems on a RIP internetwork.

Warning

The recommended method for enabling file tracing for RIP is to use the Netsh tool rather than by changing the registry settings shown here. For more information, see “Netsh ras context” earlier in this document.

Tracing consumes system resources and must be used sparingly. After you obtain the trace or identify the problem, immediately disable tracing. In particular, do not leave tracing enabled on multiprocessor computers.

The following table lists the registry entries for configuring file tracing for RIP v2.

Entries for Enabling Tracing Under the IPRIP2 Registry Key

Entry Task

EnableFileTracing

Configuring file tracing:

  • 1 enables file tracing for the RIP v2 routing protocol.

  • 0 disables file tracing for the RIP v2 routing protocol. File tracing is disabled by default.

FileDirectory

Configuring location of the file tracing log file:

  • By default, the location of the tracing file is set to %windir%\Tracing\IPRIP2.LOG file.

  • To set a new default location, you can right-click this entry and then specify an alternative for %windir%\Tracing. You cannot change the log file name.

FileTracingMask

Configuring how much tracing information is saved in the log file:

  • By default, the level for file tracing is set to 0xFFFF0000, which is the maximum level.

  • To set a new default level for file tracing, you can right-click this entry and then specify an alternative level in the range from 0 to 0xFFFF0000.

MaxFileSize

Configuring maximum log file size:

  • By default, the maximum size of the log file is set to 0x100000, which is 1 MB.

  • To set a new maximum size for the log file, you can right-click this entry and then specify an alternative maximum size.

OSPF

Registry path HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Tracing

Version Windows Server 2003 and Windows 2000 Server

Configuring entries under the OSPF key enables file tracing, which logs tracing information about the OSPF routing protocol to files so that you can troubleshoot problems on an OSPF internetwork.

Warning

The recommended method for enabling file tracing for OSPF is to use the Netsh tool rather than by changing the registry settings shown here. For more information, see “Netsh ras context” earlier in this document.

Tracing consumes system resources and must be used sparingly. After you obtain the trace or identify the problem, immediately disable tracing. In particular, do not leave tracing enabled on multiprocessor computers.

The following table lists the registry entries for configuring file tracing for OSPF.

Entries for Enabling Tracing Under the OSPF Registry Key

Entry Task

EnableFileTracing

Configuring file tracing:

  • 1 enables file tracing for the OSPF routing protocol.

  • 0 disables file tracing for the OSPF routing protocol.

FileDirectory

Configuring location of the file tracing log file:

  • By default, the location of the tracing file is set to %windir%\Tracing\OSPF.LOG file.

  • To set a new default location, you can right-click this entry and then specify an alternative for %windir%\Tracing. You cannot change the log file name.

FileTracingMask

Configuring how much tracing information is saved in the log file:

  • By default, the level for file tracing is set to 0xFFFF0000, which is the maximum level.

  • To set a new default level for file tracing, you can right-click this entry and then specify an alternative level in the range from 0 to 0xFFFF0000.

MaxFileSize

Configuring maximum log file size:

  • By default, the maximum size of the log file is set to 0x100000, which is 1 MB.

  • To set a new maximum size for the log file, you can right-click this entry and then specify an alternative maximum size.

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Tcpip\Parameters

The Routing and Remote Access service uses several TCP/IP registry keys. The entries included in this section are those that are specific to unicast IPv4 routing. For additional entries for TCP/IP, see TCP/IP Tools and Settings.

The following registry entries are located under the registry key HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters.

DisableIPSourceRouting

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000, Windows NT 4.0, or Windows 98

This entry determines whether IP source routing is enabled. IP source routing lets the sender use a tool, such as Tracert.exe or Ping.exe, to specify the exact route that an IP packet takes through successive IP routers across the IP internetwork.

Source routing is not typically implemented on IP internetworks. Instead, IP routing is typically handled through routing decisions made by source hosts and IP routers based on entries in their respective local routing tables. Although you might want to use IP source routing in network testing and debugging situations, disable IP source routing as soon as you are finished testing or troubleshooting.

Values for DisableIPSourceRouting

Value Setting

0

Forwards all packets.

1 (Default)

Does not forward source routed packets.

2

Drops all incoming source routed packets.

EnableDeadGWDetect

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000, Windows NT 4.0, or Windows 98

This entry determines whether dead gateway detection is enabled. With EnableDeadGWDetect enabled, TCP can detect a failure of the default gateway and inform IP to switch to a different default gateway if one is available. A default gateway is an IP router on the local subnet that can forward traffic to destinations located beyond the local subnet.

For optimal security, setting EnableDeadGWDetect to 0 is recommended in order to prevent an attacker from forcing this computer to switch to an unintended default gateway.

Values for EnableDeadGWDetect

Value Setting

0

Disables dead gateway detection.

1 (default)

Enables dead gateway detection.

EnableFastRouteLookup

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows 2000 Server

This entry, which is created only if the Routing and Remote Access Service is enabled, determines whether the fast route lookup feature is enabled. Fast route lookups make route searches faster but use a significant amount of non-paged pool memory. Fast route lookups are used only if the computer runs Windows Server 2003 or Windows 2000 Server and contains at least 64 MB of memory.

Values for EnableFastRouteLookup

Value Setting

0 (default)

Disables fast route lookup.

1

Enables fast route lookup.

EnableICMPRedirects

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000

This entry determines whether a computer running Windows Server 2003 TCP/IP, or another version of Windows TCP/IP that supports Internet Control Message Protocol (ICMP) Redirect messages, updates its routing table in response to ICMP Redirect messages sent to it by a network device such as a router. An ICMP Redirect message instructs the computer to direct IP packets for the recipient along a different route. A computer running Windows Server 2003 accepts ICMP redirection messages from any node in the route between it and the destination computer.

You must restart the computer for changes to this entry to take effect.

Values for EnableICMPRedirects

Value Setting

0

Disables updates of the IP routing table in response to ICMP Redirect messages.

1 (default)

Enables updates of the IP routing table in response to ICMP Redirect messages.

EnablePMTUBHDetect

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000

This entry determines whether TCP tries to detect black hole routers during the path maximum transmission unit (PMTU) discovery process. For performance reasons, PMTU black hole router detection is disabled by default for TCP/IP in Windows Server 2003, Windows XP, and Windows 2000.

When PMTU black hole router detection is enabled, if several retransmissions of a TCP segment are not acknowledged, TCP tries sending segments with the Don’t Fragment flag in the IP header of the IP packet set to 0 (which enables an IP router to fragment the TCP segment). If, as a result, a segment with the Don’t Fragment flag set to 0 is acknowledged, the TCP maximum segment size (MSS) is decreased and the Don’t Fragment flag is set to 1 (which prevents an IP router from fragmenting the TCP segment) in subsequent segments on the connection.

Enabling PMTU black hole detection increases the maximum number of retransmissions that are performed for a specific segment and can therefore decrease overall performance.

This registry entry is not present by default and must be added by using the Registry Editor tool, and then restarting Windows.

Values for EnablePMTUBHDetect

Value Setting

0 (default)

Disables PMTU black hole detection.

1

Enables PMTU black hole detection.

EnablePMTUDiscovery

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000

This entry determines whether TCP uses a fixed, default maximum transmission unit (MTU) of 576 bytes for all connections that are not to destinations on a locally attached subnet, or alternatively, whether TCP attempts to detect the actual path maximum transmission unit (PMTU).

When PMTU Discovery is enabled, TCP attempts to discover the MTU, or largest packet size, allowed by the path to a remote host. By discovering the PMTU and limiting TCP segments to this size, TCP can eliminate fragmentation at IP routers along the path that connect networks with different MTUs. Fragmentation reduces TCP throughput and increases network congestion.

By default, this entry applies to all interfaces. However, you can use the Registry Editor to change the default value of this entry for one or more interfaces on this computer.

For optimal security, set EnablePMTUDiscovery to 0 in order to prevent an attacker from forcing the MTU value to a very small value and thereby overworking the TCP/IP stack.

This registry entry is not present by default and must be added by using the Registry Editor tool, and then restarting Windows.

Values for EnablePMTUDiscovery

Value Setting

0

Causes TCP to use an MTU of 576 bytes for all connections to nodes outside of the local subnet.

1 (default)

Enables TCP to attempt to discover the actual MTU of the path to a remote node.

IPEnableRouter

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000

This entry determines whether IP forwarding is enabled on a computer on which the Routing and Remote Access service is not enabled. As the name IPEnableRouter indicates, manually enabling this entry enables the computer to act as a router. That is, when IP forwarding is enabled, the computer can route IP packets that it receives to nodes on all networks to which it is connected.

Values for IPEnableRouter

Value Setting

0 (default)

Disables IP forwarding. The host does not route IP packets.

1

Enables IP forwarding. The host routes IP packets.

A server running the Routing and Remote Access service does not use the IPEnableRouter registry entry; this entry is therefore, correctly, set to 0.

If you want to enable IP forwarding on a workstation or on a server on which you do not want to enable the Routing and Remote Access service, you must use the Registry Editor to set the IPEnableRouter registry entry to 1. Typically, a workstation or multi-homed server that is not a router should not be configured to forward packets between networks, unless it is acting as a firewall.

IPEnableRouterBackup

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

Version Windows Server 2003, Windows XP, Windows 2000

This entry stores a backup copy of the value that was stored in the IPEnableRouter entry when the operating system of the computer was most recently installed or upgraded. The computer uses this entry to:

  • Determine whether the value of IPEnableRouter has changed

  • Restore the previous value if the value of IPEnableRouter has changed

Do not delete this entry or change its value. If you do, you might disable the IP packet routing mechanism.

There is no default value for this entry. The computer enters a value when you install or upgrade Windows.

Values for IPEnableRouterBackup

Value Setting

0

Disables storing a backup copy of the value stored in IPEnableRouter.

1

Enables storing a backup copy of the value stored in IPEnableRouter.

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces\interfaceGUID

The Routing and Remote Access service uses several TCP/IP registry keys for interfaces. The entries included in this section are specific to unicast IPv4 routing features supported by the Routing and Remote Access service. For additional entries for TCP/IP interfaces, see TCP/IP Tools and Settings.

The following registry entries are located under the registry key HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces\interfaceGUID.

Before you can use the registry editor to configure a registry entry located under this key, you must first determine the globally unique identifier (GUID) for the TCP/IP interface that you want to configure.

To determine the GUID for a specific TCP/IP interface

  1. Open Network Connections, and then note the name of the LAN connection, such as “Local Area Connection.”

  2. Open the Registry Editor (click Start, click Run, type regedit.exe, and then click OK).

  3. In the tree view (the left pane), expand the following key: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control \Network\{4D36E972-E325-11CE-BFC1-08002BE10318}

  4. Under this key are one or more keys for the globally unique identifiers (GUIDs) corresponding to the installed LAN connections on this computer. Each of these GUID keys has a Connection subkey. Open each of the GUID\Connection keys, and then look for the Name setting whose value matches the name of your LAN connection from step 1.

  5. When you have found the GUID\Connection key that contains the Name setting that matches the name of your LAN connection, write down, take a screenshot, or otherwise note the GUID value.

  6. Use the tree view to open the following key: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip \Parameters\Interfaces\interfaceGUID.

  7. Right-click the interfaceGUID key you want in the tree view, and then view or configure that entry.

PerformRouterDiscovery

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces\interfaceGUID

Version Windows Server 2003, Windows XP, Windows 2000, Windows 98

Note

The recommended method for configuring router discovery on a host running Windows Server 2003 or Windows XP is to set DHCP options, instead of modifying the registry.

The PerformRouterDiscovery entry determines whether Windows TCP/IP attempts to perform ICMP router discovery as a host, as defined by RFC 1256, on this interface. Enabling router discovery on a client or non-router server enables that computer to listen for Router Advertisement messages from an IP router, to send Router Solicitation messages to discover active routers on its subnet, and to switch to another router if the first router fails or the network administrator changes router preferences.

Windows does not add the PerformRouterDiscovery entry to the registry. To enable ICMP router discovery, you must add this entry (as type REG_DWORD) and set its value to one of the options listed in the following table.

Values for Enabling PerformRouterDiscovery

Value Setting

0

Disables router discovery.

1

Enables router discovery.

2 (default)

Enables DHCP-controlled router discovery.

The value 2 is the default, but it is used only if the DHCP options Perform Router Discovery and Router Solicitation Address are configured. The DHCP Perform Router Discovery option is DHCP scope option 31 and the DHCP Router Solicitation Address option is DHCP scope option 32. Both are configured by using the DHCP snap-in. For more information, see “DHCP Options” in DHCP Tools and Settings.

For ICMP router discovery to take place, in addition to configuring PerformRouterDiscovery on a client or non-router server, the following requirements must also be met:

  • A Routing and Remote Access router on the same subnet must be configured for ICMP router discovery, to enable it to send Router Advertisement messages to all hosts on the subnet. For more information, see “Netsh routing IP routerdiscovery context” or “ICMP router discovery options” earlier in this technical reference.

  • The entry SolicitationAddressBcast must be added to the registry, and its value configured. For more information, see “SolicitationAddressBcast” later in this technical reference.

PerformRouterDiscoveryBackup

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces\interfaceGUID

Version Windows Server 2003, Windows XP, Windows 2000

This entry stores a backup copy of the value stored in the PerformRouterDiscovery entry (if an entry for PerformRouterDiscovery has been configured). The computer uses this entry only when IP routing is enabled and the value of PerformRouterDiscovery is 0.

Warning

This entry is used internally by the Windows operating system. Do not delete this entry or change its value. If you do, you might disable the router discovery mechanism.

There is no default value for this entry.

Values for PerformRouterDiscoveryBackup

Value Setting

0

Disables storing a backup copy of the value stored in PerformRouterDiscovery.

1

Enables storing a backup copy of the value stored in PerformRouterDiscovery.

SolicitationAddressBcast

Registry path HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces\interfaceGUID

Version Windows Server 2003, Windows XP, Windows 2000, Windows 98

Note

The recommended method for configuring router discovery on a host running Windows Server 2003 or Windows XP is to set DHCP options, instead of modifying the registry.

The SolicitationAddressBcast entry, as defined by RFC 1256, determines whether Windows sends router discovery messages as broadcasts rather than as multicasts on this interface. Enabling router discovery on a client or non-router server computer lets that computer listen for Router Advertisement messages from a router and lets that computer send Router Solicitation messages to discover active routers on its subnet and to discover the unavailability of previously available routers that are currently down.

Windows does not add the SolicitationAddressBcast entry to the registry. To specify how router solicitation messages are sent, you must add this entry (as type REG_DWORD) and set its value to one of the options listed in the following table.

Values for Enabling SolicitationAddressBcast

Value Setting

0 (Default)

Enables the sending of router solicitation messages to the all-routers multicast group (224.0.0.2). This is the recommended option.

1

Enables the sending of router solicitation messages as limited broadcast messages by using the IP address 255.255.255.255.

The value 1 is not recommended because it causes all clients to broadcast router solicitation messages, creating a large volume of unnecessary broadcast traffic that must be processed by other clients.

For ICMP router discovery to take place, in addition to configuring the value for this entry, the following requirements must also be met:

  • A Routing and Remote Access router on the same subnet must be configured for ICMP router discovery, to enable it to send Router Advertisement messages to all hosts on the subnet. For more information, see “Netsh routing IP routerdiscovery context” or “ICMP router discovery options” earlier in this technical reference.

  • The entry PerformRouterDiscovery must be added to the registry, and its value configured. For more information, see “PerformRouterDiscovery” earlier in this technical reference.

Note that PerformRouterDiscovery option 2 enables DHCP-controlled router discovery, but only if the DHCP options Perform Router Discovery and Router Solicitation Address are configured.

The following resources contain additional information that is relevant to this section.